sPHENIX discussion of software

["Huang, Jin" <jhuang AT bnl.gov>]

Also available on agenda: https://indico.bnl.gov/conferenceDisplay.py?confId=1179

 

 

Announcement:

sPHENIX simulation workfest in July @ SBU. Please email Jin for your dates of attendance and interest:

https://indico.bnl.gov/conferenceDisplay.py?confId=1237

 

 

Mike - New evaluator

 

Page 2-3: link DST object for forward backward connection between truth particle and tracks/cal. clusters.

Jin: Are there lower level connections available? 

Mike: yes, example includes truth to SvxCluster or to Calorimeter towers, which are also supported on DST

 

Page 6: object ID are used in the association.

Jin: does every object has an ID rather than relying on the location on the array. Example is PHG4VtxPoint in the PHG4TruthInfoContainer does not carry an ID, rather than relying on the index in the PHG4TruthInfoContainer::vtxmap.

Mike & Chris: it should be our policy to do so.

 

Action Item [Jin]: add index to PHG4VtxPoint

 

Page 8:

Chris, Jamie & Mike: ROOT imposed a problem to use BOOST in the IO objects. We can still use BOOST in the processing modules.

 

End discussion:

 

Dave: noticed similarity between the link operation and SQL operation. Can we borrow anything?

Mike: portability issue? Current implementation is mutable in the future. We can check with SQL and see what we can borrow over.

 

Jin: Are we linking to the primary particle or both primary and secondary particles?

Mike: Currently primary particles.

 

Chris: we may need add noise hits association.

Mike: used NULL ID for this purpose. 

 

Jin: plan on availability. We can test it out in user module for the HCal Analysis

Mike: will keep interested party notified.

 

 

Jin - framework

 

Page 5:

Tony: how git handle conflicting editing?

John L., Martin and Jin: Git handle most cases to merge parallel edits to different part of the files. For conflictions between parallel developments, we need to sort out through

  1) process Git Pull request quickly so updates are applied to master branch so other developers can see. 

  2) unavoidable for software development for internal coordination.

 

Page 8-9:

Peter: Instead of using a single array of property ID -> union of (float, int, uint), one could consider using three separate maps of  property ID -> float, ID-> int and ID -> uint. Goal is to improve clarity.

Jin: It is a viable option. No strong preference in either way.

Chris: union structure as implemented now is preferred to avoid duplicated entries of same property ID to different types.

 

Action Item [Jin]: Try 64 bit storage of the property and whether ROOT can compress them to reasonable size.

 

 

Eliton – tuning scintillation model for EMCal and HCals: Birk's correction VS Geant4 max step length

 

Page 3:

John H. : do we know Birk's constant for Hcal scintillator? Need to calibration with some tests.

Jin: we implemented the standard polystyrene's Birk's constant from M.Hirschberg et al., IEEE Trans. Nuc. Sci. 39 (1992) 511 (same one used in Geant4 by default). It is our best guess. Test value are welcomed and we can plug in later.

 

Page 4:

Jin: btw, we should expect the light yield be lower at smaller step length.

 

Page 5-7:

Peter: we can use histogram mean rather than Gauss fit, since no peak range is pre-selected before fit.

 

Page 12:

Peter: histogram dE/dx vs (Birk's effect) to check the correlation.

 

Action Item [Eliton] : Try histogram dE/dx vs (Birk's effect) to check the correlation.

 

Chris: an easier test would be taking a block of very large single piece of scintillator for this study.

 

Action Item [Eliton]: finish this test and implement proper step length limit for both EMCal and HCals.

 

 

______________________________

 

Jin HUANG

 

Brookhaven National Laboratory

Physics Department, Bldg 510 C

Upton, NY 11973-5000

 

Office: 631-344-5898

Cell:   757-604-9946

______________________________

 

From: Huang, Jin
Sent: Monday, June 22, 2015 2:45 PM
To: phenix-nextmid-l; 'sphenix-software-l AT lists.bnl.gov'
Subject: [reminder] Tue Jun 23 1-3PM sPHENIX simulation meetings @ 2-160

 

Dear All,

 

This is reminder that we will resume our weekly sPHENIX simulation meeting Tue Jun 23 1-3PM, after two weeks of interruptions. For residents at BNL, we will meet in the 2-160 conference room.

 

Currently on agenda,

·       Mike will introduce the new evaluator for sPHENIX

·       Chris and I will discuss organization of software repository

·       Eliton will discuss the scintillation light model for the calorimeters.

·       New topics are welcomed too.  

The agenda is at https://indico.bnl.gov/conferenceDisplay.py?confId=1179 , for which the modification password is 1008

 

To join via Browser:

https://bluejeans.com/377457648/browser

 

To join via Phone:

1) Dial:

+1 408 740 7256

+1 888 240 2560(US Toll Free)

+1 408 317 9253(Alternate Number)

(see all numbers - http://bluejeans.com/numbers)

2) Enter Conference ID: 377457648

 

Cheers,

 

Jin

 

______________________________

 

Jin HUANG

 

Brookhaven National Laboratory

Physics Department, Bldg 510 C

Upton, NY 11973-5000

 

Office: 631-344-5898

Cell:   757-604-9946

______________________________